idnits 2.17.1 draft-ietf-cdni-footprint-capabilities-semantics-02.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (February 14, 2014) is 3724 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- ** Obsolete normative reference: RFC 5226 (Obsoleted by RFC 8126) == Outdated reference: A later version (-14) exists of draft-ietf-cdni-framework-09 == Outdated reference: A later version (-27) exists of draft-ietf-cdni-logging-09 == Outdated reference: A later version (-21) exists of draft-ietf-cdni-metadata-04 Summary: 1 error (**), 0 flaws (~~), 4 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 CDNI J. Seedorf 3 Internet-Draft NEC 4 Intended status: Informational J. Peterson 5 Expires: August 18, 2014 Neustar 6 S. Previdi 7 Cisco 8 R. van Brandenburg 9 TNO 10 K. Ma 11 Azuki Systems, Inc. 12 February 14, 2014 14 CDNI Request Routing: Footprint and Capabilities Semantics 15 draft-ietf-cdni-footprint-capabilities-semantics-02 17 Abstract 19 This document tries to capture the semantics of the "Footprint and 20 Capabilities Advertisement" part of the CDNI Request Routing 21 interface, i.e. the desired meaning and what "Footprint and 22 Capabilities Advertisement" is expected to offer within CDNI. The 23 discussion in this document has the goal to facilitate the choosing 24 of one or more suitable protocols for "Footprint and Capabilities 25 Advertisement" within CDNI Request Routing. 27 Requirements Language 29 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 30 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 31 document are to be interpreted as described in RFC 2119 [RFC2119]. 33 Status of This Memo 35 This Internet-Draft is submitted in full conformance with the 36 provisions of BCP 78 and BCP 79. 38 Internet-Drafts are working documents of the Internet Engineering 39 Task Force (IETF). Note that other groups may also distribute 40 working documents as Internet-Drafts. The list of current Internet- 41 Drafts is at http://datatracker.ietf.org/drafts/current/. 43 Internet-Drafts are draft documents valid for a maximum of six months 44 and may be updated, replaced, or obsoleted by other documents at any 45 time. It is inappropriate to use Internet-Drafts as reference 46 material or to cite them other than as "work in progress." 48 This Internet-Draft will expire on August 18, 2014. 50 Copyright Notice 52 Copyright (c) 2014 IETF Trust and the persons identified as the 53 document authors. All rights reserved. 55 This document is subject to BCP 78 and the IETF Trust's Legal 56 Provisions Relating to IETF Documents 57 (http://trustee.ietf.org/license-info) in effect on the date of 58 publication of this document. Please review these documents 59 carefully, as they describe your rights and restrictions with respect 60 to this document. Code Components extracted from this document must 61 include Simplified BSD License text as described in Section 4.e of 62 the Trust Legal Provisions and are provided without warranty as 63 described in the Simplified BSD License. 65 Table of Contents 67 1. Introduction and Scope . . . . . . . . . . . . . . . . . . . 2 68 2. Design Decisions for Footprint and Capabilities . . . . . . . 4 69 2.1. Advertising Limited Coverage . . . . . . . . . . . . . . 4 70 2.2. Capabilities and Dynamic Data . . . . . . . . . . . . . . 5 71 2.3. Advertisement versus Queries . . . . . . . . . . . . . . 6 72 2.4. Avoiding or Handling 'cheating' dCDNs . . . . . . . . . . 7 73 2.5. Focus on Main Use Cases may Simplify Things . . . . . . . 7 74 3. Main Use Case to Consider . . . . . . . . . . . . . . . . . . 8 75 4. Semantics for Footprint Advertisement . . . . . . . . . . . . 8 76 5. Semantics for Capabilities Advertisement . . . . . . . . . . 10 77 6. Negotiation of Support for Optional Types of 78 Footprint/Capabilities . . . . . . . . . . . . . . . . . . . 13 79 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 13 80 7.1. Footprint Sub-Registry . . . . . . . . . . . . . . . . . 14 81 7.2. Protocol Sub-Registry . . . . . . . . . . . . . . . . . . 14 82 7.3. Redirection Mode Sub-Registry . . . . . . . . . . . . . . 15 83 8. Security Considerations . . . . . . . . . . . . . . . . . . . 15 84 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 15 85 9.1. Normative References . . . . . . . . . . . . . . . . . . 15 86 9.2. Informative References . . . . . . . . . . . . . . . . . 16 87 Appendix A. Acknowledgment . . . . . . . . . . . . . . . . . . . 16 88 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 17 90 1. Introduction and Scope 92 The CDNI working group is working on a set of protocols to enable the 93 interconnection of multiple CDNs to a CDN federation. This CDN- 94 federation should serve multiple purposes, as discussed in [RFC6770], 95 for instance, to extend the reach of a given CDN to areas in the 96 network which are not covered by this particular CDN. 98 The goal of this document is to achieve a clear understanding in the 99 CDNI WG about the semantics associated with the CDNI Request Routing 100 Footprint & Capabilities Advertisement Interface (from now on 101 referred to as FCI), in particular the type of information a 102 downstream CDN 'advertises' regarding its footprint and capabilities. 103 To narrow down undecided aspects of these semantics, this document 104 tries to establish a common understanding of what the FCI should 105 offer and accomplish in the context of CDN Interconnection. 107 It is explicitly outside the scope of this document to decide on 108 specific protocols to use for the FCI. 110 General assumptions in this document: 112 o The CDNs participating in the CDN federation have already 113 performed a boot strap process, i.e., they have connected to each 114 other, either directly or indirectly, and can exchange information 115 amongst each other. 117 o The uCDN has received footprint and/or capability advertisements 118 from a set of dCDNs. Footprint advertisement and capability 119 advertisement need not use the same underlying protocol. 121 o The upstream CDN (uCDN) receives the initial request-routing 122 request from the endpoint requesting the resource. 124 The CDNI Problem Statement [RFC6707] describes footprint and 125 capabilities advertisement as: "[enabling] a Request Routing function 126 in an Upstream CDN to query a Request Routing function in a 127 Downstream CDN to determine if the Downstream CDN is able (and 128 willing) to accept the delegated Content Request". In addition, the 129 RFC says "the CDNI Request Routing interface is also expected to 130 enable a downstream CDN to provide to the upstream CDN (static or 131 dynamic) information (e.g. resources, footprint, load) to facilitate 132 selection of the downstream CDN by the upstream CDN request routing 133 system when processing subsequent content requests from User Agents". 134 It thus considers "resources" and "load" as capabilities to be 135 advertised by the downstream CDN. 137 The range of different footprint definitions and possible 138 capabilities is very broad. Attempting to define a comprehensive 139 advertisement solution quickly becomes intractable. The CDNI 140 requirements draft [I-D.ietf-cdni-requirements] lists the specific 141 requirements for the CDNI Footprint & Capabilities Advertisement 142 Interface in order to disambiguate footprints and capabilities with 143 respect to CDNI. This document attempts to distill the apparent 144 common understanding of what the terms 'footprint' and 'capabilities' 145 mean in the context of CDNI, and detail the semantics of the 146 footprint advertisement mechanism and the capability advertisement 147 mechanism. 149 2. Design Decisions for Footprint and Capabilities 151 A large part of the difficulty in discussing the FCI lies in 152 understanding what exactly is meant when trying to define footprint 153 in terms of "coverage" or "reachability." While the operators of 154 CDNs pick strategic locations to situate caches, a cache with a 155 public IPv4 address is reachable by any endpoint on the Internet 156 unless some policy enforcement precludes the use of the cache. 158 Some CDNs aspire to cover the entire world, which we will henceforth 159 call global CDNs. The footprint advertised by such a CDN in the CDNI 160 environment would, from a coverage or reachability perspective, 161 presumably cover all prefixes. Potentially more interesting for CDNI 162 use cases, however, are CDNs that claim a more limited coverage, but 163 seek to federate with other CDNs in order to create a single CDN 164 fabric which shares resources. 166 Futhermore, not all capabilities need be footprint restricted. 167 Depending upon the use case, the optimal semantics of "footprints 168 with capability attributes" vs. "capabilities with footprint 169 restrictions" are not clear. 171 The key to understanding the semantics of footprint and capability 172 advertisement lies in understand why a dCDN would advertise a limited 173 coverage area, and how a uCDN would use such advertisements to decide 174 among one of several dCDNs. The following section will discuss some 175 of the trade-offs and design decisions that need to be decided upon 176 for the CDNI FCI. 178 2.1. Advertising Limited Coverage 180 The basic use case that would motivate a dCDN to advertise a limited 181 coverage is that the CDN was built to cover only a particular portion 182 of the Internet. For example, an ISP could purpose-build a CDN to 183 serve only their own customers by situating caches in close 184 topological proximity to high concentrations of their subscribers. 185 The ISP knows the prefixes it has allocated to end users and thus can 186 easily construct a list of prefixes that its caches were positioned 187 to serve. 189 When such a purpose-built CDN joins a federation, however, and 190 advertises its footprint to a uCDN, the original intended coverage of 191 the CDN might not represent its actual value to the federation of 192 CDNs. Consider an ISP-A and ISP-B that both field their own CDNs, 193 which they federate through CDNI. A given user E, who is customer of 194 ISP-B, might happen to be topologically closest to a cache fielded by 195 ISP-A, if E happens to live in a region where ISP-B has few customers 196 and ISP-A has many. In this case, should ISP-A's CDN "cover" E? If 197 ISP-B's CDN has a failure condition, should the uCDN understand that 198 ISP-A's caches are potentially available back-ups - and if so, how 199 does ISP-A advertise itself as a "standby" for E? What about the 200 case where CDNs advertising to the same uCDN express overlapping 201 coverage (for example, a federation mixing global and limited CDNs)? 203 The answers to these questions greatly depend on how much information 204 we want the uCDN to use to make a selection of a dCDN. If a uCDN has 205 three dCDNs to choose from that "cover" the IP address of user E, 206 obviously the uCDN might be interested to know how optimal the 207 coverage is from each of the dCDNs - coverage need not be binary, 208 either provided or not provided. dCDNs could advertise a coverage 209 "score," for example, and provided that they all reported scores 210 fairly on the same scale, uCDNs could use that to make their 211 topological optimality decision. Alternatively, dCDNs could for 212 their footprint advertise the IP addresses of their caches rather 213 than prefix "coverage," and let the uCDN decide for itself (based on 214 its own topological intelligence) which dCDN has better resources to 215 serve a given user. 217 In summary, the semantics of advertising footprint depend on whether 218 such qualitative metrics for expressing footprint (such as the 219 coverage 'score' mentioned above) should be part of the CDNI FCI, or 220 if it should focus just on 'binary' footprint. 222 2.2. Capabilities and Dynamic Data 224 In cases where the apparent footprint of dCDNs overlaps, uCDNs might 225 also want to rely on a host of other factors to evaluate the 226 respective merits of dCDNs. These include facts related to the 227 caches themselves, to the network where the cache is deployed, to the 228 nature of the resource sought and to the administrative policies of 229 the respective networks. 231 In the absence of network-layer impediments to reaching caches, the 232 choice to limit coverage is necessarily an administrative policy. 233 Much policy must be agreed upon before CDNs can merge into 234 federations, including questions of membership, compensation, volumes 235 and so on. A uCDN certainly will factor these sorts of 236 considerations into its decision to select a dCDN, but there is 237 probably little need for dCDNs to actually advertise them through an 238 interface - they will be settled out of band as a precondition for 239 federating. 241 Other facts about the dCDN would be expressed through the interface 242 to the uCDN. Some capabilities of a dCDN are static, and some are 243 highly dynamic. Expressing the total storage built into its caches, 244 for example, changes relatively rarely, whereas the amount of storage 245 in use at any given moment is highly volatile. Network bandwidth 246 similarly could be expressed as either total bandwidth available to a 247 cache, or based on the current state of the network. A cache may at 248 one moment lack a particular resource in storage, but have it the 249 next. 251 The semantics of the capabilities interface will depend on how much 252 of the dCDN state needs to be pushed to the uCDN and qualitatively 253 how often that information should be updated. 255 2.3. Advertisement versus Queries 257 In a federated CDN environment, each dCDN shares some of its state 258 with the uCDN, which the uCDN uses to build a unified picture of all 259 of the dCDNs available to it. In architectures that share detailed 260 capability information, the uCDN could basically perform the entire 261 request-routing intelligence down to selecting a particular cache 262 before sending the request to the dCDN (note that within the current 263 CDNI WG scope, such direct selection of specific caches by the uCDN 264 is out of scope). However, when the uCDN must deal with many 265 potential dCDNs, this approach does not scale. Especially as CDNs 266 scale up from dozens or hundreds of caches to thousands or tens of 267 thousands, the volume of updates to footprint and capability may 268 become onerous. 270 Were the volume of updates to exceed the volumes of requests to the 271 uCDN, it might make more sense for the uCDN to query dCDNs upon 272 receiving requests (as is the case in the recursive redirection mode 273 described in [I-D.ietf-cdni-framework]), instead of receiving 274 advertisements and tracking the state of dCDNs itself. The advantage 275 of querying dCDNs would be that much of the dynamic data that dCDNs 276 cannot share with the uCDN would now be factored into the uCDN's 277 decision. dCDNs need not replicate any state to the uCDN - uCDNs 278 could effectively operate in a stateless mode. 280 The semantics of both footprint and capability advertisement depend 281 on the service model here: are there cases where a synchronous query/ 282 response model would work better for the uCDN decision than a state 283 replication model? 285 2.4. Avoiding or Handling 'cheating' dCDNs 287 In a situation where more than one dCDN is willing to serve a given 288 end user request, it might be attractive for a dCDN to 'cheat' in the 289 sense that the dCDN provides inaccurate information to the uCDN in 290 order to convince the uCDN to select it opposed to 'competing' dCDNs. 291 It could therefore be desirable to take away the incentive for dCDNs 292 to cheat (in information advertised) as much as possible. One option 293 here is to make the information the dCDN advertises somehow 294 verifiable for the uCDN. One the other hand, a cheating dCDN might 295 be avoided or handled by the fact that there will be strong 296 contractual agreements between a uCDN and a dCDN, so that a dCDN 297 would risk severe penalties or legal consequences when caught 298 cheating. 300 Overall, it seems that information a dCDN advertises should (in the 301 long run) be somehow qualitatively verifiable by the uCDN, though 302 possibly through non-real-time out-of-band audits. It is probably an 303 overly strict requirement to mandate that such verification be 304 possible "immediately", i.e. during the request routing process 305 itself. If the uCDN can detect a cheating dCDN at a later stage, it 306 should suffice for the uCDN to "de-incentivize" cheating because it 307 would negatively affect the long-term business relationship with a 308 particular dCDN. 310 2.5. Focus on Main Use Cases may Simplify Things 312 To narrow down semantics for "footprint" and "capabilities" in the 313 CDNI context, it can be useful to initially focus on key use cases to 314 be addressed by the CDNI WG that are to be envisioned the main 315 deployments in the foreseeable future. In this regard, a main 316 realistic use case is the existence of ISP-owned CDNs, which 317 essentially cover a certain operator's network. At the same time, 318 however, the possibility of overlapping footprints should not be 319 excluded, i.e. the scenario where more than one dCDN claims it can 320 serve a given end user request. The ISPs may also choose to federate 321 with a fallback global CDN. 323 It seems reasonable to assume that in most use cases it is the uCDN 324 that makes the decision on selecting a certain dCDN for request 325 routing based on information the uCDN has received from this 326 particular dCDN. It may be assumed that 'cheating' CDNs will be 327 dealt with via means outside the scope of CDNI and that the 328 information advertised between CDNs is accurate. In addition, 329 excluding the use of qualitative information (e.g., cache proximity, 330 delivery latency, cache load) to predict the quality of delivery 331 would further simplify the use case allowing it to better focus on 332 the basic functionality of the FCI. 334 3. Main Use Case to Consider 336 Focusing on a main use case that contains a simple (yet somewhat 337 challenging), realistic, and generally imaginable scenario can help 338 in narrowing down the requirements for the CDNI FCI. To this end, 339 the following (simplified) use case can help in clarifying the 340 semantics of footprint and capabilities for CDNI. In particular, the 341 intention of the use case is to clarify what information needs to be 342 exchanged on the CDNI FCI, what types of information need to be 343 supported in a mandatory fashion (and which should be considered 344 optional), and what types of information need to be updated with 345 respect to a priori established CDNI contracts. 347 In short, one can imagine the following use case: A given uCDN has 348 several dCDNs. It selects one dCDN for delivery protocol A and 349 footprint 1 and another dCDN for delivery protocol B and footprint 1. 350 The dCDN that serves delivery protocol B has a further, transitive 351 (level-2) dCDN, that serves delivery protocol B in a subset of 352 footprint 1 where the first-level dCDN cannot serve delivery protocol 353 B itself. What happens if capabilities change in the transitive 354 level-2 dCDN that might affect how the uCDN selects a level-1 dCDN 355 (e.g. in case the level-2 dCDN cannot serve delivery protocol B 356 anymore)? How will these changes be conveyed to the uCDN? In 357 particular, what information does the uCDN need to be able to select 358 a new first-level dCDN, either for all of footprint 1 or only for the 359 subset of footprint 1 that the transitive level-2 dCDN served on 360 behalf of the first-level dCDN? 362 4. Semantics for Footprint Advertisement 364 Roughly speaking, "footprint" can be defined as "ability and 365 willingness to serve" by a downstream CDN. However, in addition to 366 simple "ability and willingness to serve", the uCDN may wish to have 367 additional information to make a dCDN selection decision, e.g., "how 368 well" a given dCDN can actually serve a given end user request. The 369 "ability and willingness" to serve should be distinguished from the 370 subjective qualitative measurement of "how well" it was served. One 371 can imagine that such additional information is implicitly associated 372 with a given footprint, e.g. due to contractual agreements (e.g. 373 SLAs), business relationships, or perceived dCDN quality in the past. 374 As an alternative, such additional information could also be 375 explicitly tagged along with the footprint. 377 It is reasonable to assume that a significant part of the actual 378 footprint advertisement will happen in contractual agreements between 379 participating CDNs, i.e. prior to the advertisement phase using the 380 CDNI FCI. The reason for this assumption is that any contractual 381 agreement is likely to contain specifics about the dCDN coverage 382 (i.e. the dCDN footprint) the contractual agreement applies to. In 383 particular, additional information to judge the delivery quality 384 associated with a given dCDN footprint might be defined in 385 contractual agreements (i.e. outside of the CDNI FCI). Further, one 386 can assume that dCDN contractual agreements about the delivery 387 quality associated with a given footprint will probably be based on 388 high-level aggregated statistics (i.e. not too detailed). 390 Given that a large part of footprint advertisement will actually 391 happen in contractual agreements, the semantics of CDNI footprint 392 advertisement refer to answering the following question: what exactly 393 still needs to be advertised by the CDNI FCI? For instance, updates 394 about temporal failures of part of a footprint can be useful 395 information to convey via the CDNI request routing interface. Such 396 information would provide updates on information previously agreed in 397 contracts between the participating CDNs. In other words, the CDNI 398 FCI is a means for a dCDN to provide changes/updates regarding a 399 footprint it has prior agreed to serve in a contract with a uCDN. 401 Generally speaking, one can imagine two categories of footprint to be 402 advertised by a dCDN: 404 o Footprint could be defined based on "coverage/reachability", where 405 coverage/reachability refers to a set of prefixes, a geographic 406 region, or similar boundary. The dCDN claims that it can cover/ 407 reach 'end user requests coming from this footprint'. 409 o Footprint could be defined based on "resources", where resources 410 refers to surrogates/caches a dCDN claims to have (e.g., the 411 location of surrogates/resources). The dCDN claims that 'from 412 this footprint' it can serve incoming end user requests. 414 For each of these footprint types, there are capabilities associated 415 with a given footprint, i.e. the capabilities (e.g., delivery 416 protocol, redirection mode, metadata) supported in the coverage area 417 for a "coverage/reachability" defined footprint, or the capabilities 418 of resources (e.g., delivery protocol, redirection mode, metadata 419 support) for a "resources" defined footprint. 421 It seems clear that "coverage/reachability" types of footprint must 422 be supported within CDNI. The following such types of footprint are 423 mandatory and must be supported by the CDNI FCI: 425 o List of ISO Country Codes 427 o List of AS numbers 429 o Set of IP-prefixes 430 A 'set of IP-prefixes' must be able to contain full IP addresses, 431 i.e., a /32 for IPv4 and a /128 for IPv6, and also IP prefixes with 432 an arbitrary prefix length. There must also be support for multiple 433 IP address versions, i.e., IPv4 and IPv6, in such a footprint. 435 For all of these mandatory-to-implement footprint types, footprints 436 can be viewed as constraints for delegating requests to a dCDN: A 437 dCDN footprint advertisement tells the uCDN the limitations for 438 delegating a request to the dCDN. For IP prefixes or ASN(s), the 439 footprint signals to the uCDN that it should consider the dCDN a 440 candidate only if the IP address of the request routing source falls 441 within the prefix set (or ASN, respectively). The CDNI 442 specifications do not define how a given uCDN determines what address 443 ranges are in a particular ASN. Similarly, for country codes a uCDN 444 should only consider the dCDN a candidate if it covers the country of 445 the request routing source. The CDNI specifications do not define 446 how a given uCDN determines the country of the request routing 447 source. Multiple footprint constraints are additive, i.e. the 448 advertisement of different types of footprint narrows the dCDN 449 candidacy cumulatively. 451 It addition to these mandatory "coverage/reachability" types of 452 footprint, other optional "coverage/reachability" types of footprint 453 or "resource" types of footprint may defined by future 454 specifications. To facilitate this, a clear process for specifying 455 optional footprint types in a IANA registry must be specified. This 456 includes the specification of the level of oversight necessary (e.g. 457 WG decision or expert review) for adding new optional footprints to a 458 IANA registry as well as the specification of a template regarding 459 design choices that must be captured by new optional types of 460 footprints. 462 Independent of the exact type of a footprint, a footprint might also 463 include the connectivity of a given dCDN to other CDNs that may be 464 able to serve content to users on behalf of that dCDN, to cover cases 465 where there is a transitive CDN interconnection. Further, the 466 downstream CDN must be able to express its footprint to an interested 467 upstream CDN (uCDN) in a comprehensive form, e.g., as a complete data 468 set containing the complete footprint. Making incremental updates, 469 however, to express dynamic changes in state is also desirable. 471 5. Semantics for Capabilities Advertisement 473 In general, the dCDN must be able to express its general capabilities 474 to the uCDN. These general capabilities could express if the dCDN 475 supports a given service, for instance, HTTP delivery, RTP/RTSP 476 delivery or RTMP. Furthermore, the dCDN must be able to express 477 particular capabilities for the delivery in a particular footprint 478 area. For example, the dCDN might in general offer RTMP but not in 479 some specific areas, either for maintenance reasons or because the 480 caches covering this particular area cannot deliver this type of 481 service. Hence, in certain cases footprint and capabilities are tied 482 together and cannot be interpreted independently from each other. In 483 such cases, i.e. where capabilities must be expressed on a per 484 footprint basis, it may be beneficial to combine footprint and 485 capabilities advertisement. 487 A high-level and very rough semantic for capabilities is thus the 488 following: Capabilities are types of information that allow a uCDN to 489 determine if a downstream CDN is able (and willing) to accept (and 490 properly handle) a delegated content request. In addition, 491 Capabilities are characterized by the fact that this information may 492 possibly change over time based on the state of the network or 493 caches. 495 At a first glance, several broad categories of capabilities seem 496 useful to convey via an advertisement interface, however, advertising 497 capabilities that change highly dynamically (e.g. real-time delivery 498 performance metrics, CDN resource load, or other highly dynamically 499 changing QoS information) should probably not be in scope for the 500 CDNI FCI. First, out of the multitude of possible metrics and 501 capabilities, it is hard to agree on a subset and the precise metrics 502 to be used. Second, and perhaps more importantly, it seems not 503 feasible to specify such highly dynamically changing capabilities and 504 the corresponding metrics within the CDNI charter time-frame. 506 Useful capabilities refer to information that does not change highly 507 dynamically and which in many cases is absolutely necessary to decide 508 on a particular dCDN for a given end user request. For instance, if 509 an end user request concerns the delivery of a video file with a 510 certain protocol (e.g. RTMP), the uCDN needs to know if a given dCDN 511 has the capabilitity of supporting this delivery protocol. 513 Similar to footprint advertisement, it is reasonable to assume that a 514 significant part of the actual (resource) capabilities advertisement 515 will happen in contractual agreements between participating CDNs, 516 i.e. prior to the advertisement phase using the CDNI FCI. The role 517 of capability advertisement is hence rather to enable the dCDN to 518 update a uCDN on changes since a contract has been set up (e.g. in 519 case a new delivery protocol is suddenly being added to the list of 520 supported delivery protocols of a given dCDN, or in case a certain 521 delivery protocol is suddenly not being supported anymore due to 522 failures). Capabilities advertisement thus refers to conveying 523 information to a uCDN about changes/updates of certain capabilities 524 with respect to a given contract. 526 Given these semantics, it needs to be decided what exact capabilities 527 are useful and how these can be expressed. Since the details of CDNI 528 contracts are not known at the time of this writing (and the CDNI 529 interface should probably be agnostic to these contracts anyway), it 530 remains to be seen what capabilities will be used to define 531 agreements between CDNs in practice. One implication for 532 standardization may be to initially only specify a very limited set 533 of mandatory capabilities for advertisement and have on top of that a 534 flexible data model that allows exchanging additional capabilities 535 when needed. Still, agreement needs to be found on which 536 capabilities (if any) should be mandatory among CDNs. As discussed 537 in Section 2.5, finding the concrete answers to these questions can 538 benefit from focusing on a small number of key use cases that are 539 highly relevant and contain enough complexity to help in 540 understanding what concrete capabilities are needed to facilitate CDN 541 Interconnection. 543 Under the above considerations, the following capabilities seem 544 useful as 'base' capabilities, i.e. ones that are needed in any case 545 and therefore constitute mandatory capabilities to be supported by 546 the CDNI FCI: 548 o Delivery Protocol (e.g., HTTP vs. RTMP) 550 o Acquisition Protocol (for aquiring content from a uCDN) 552 o Redirection Mode (e.g., DNS Redirection vs. HTTP Redirection as 553 discussed in [I-D.ietf-cdni-framework]) 555 o Capabilities related to CDNI Logging (e.g., supported logging 556 mechanisms) 558 o Capabilities related to CDNI Metadata (e.g., authorization 559 algorithms or support for proprietary vendor metadata) 561 It is not feasable to enumerate all the possible options for the 562 mandatory capabilities listed above (e.g., all the potential delivery 563 protocols or metadata options) or anticipate all the future needs for 564 additional capabilities. It would be unreasonable to burden the CDNI 565 FCI specification with defining each supported capability. Instead, 566 the CDNI FCI specification should define a generic protocol for 567 conveying any capability information. In this respect, it seems 568 reasonable to define a registry which initially contains the 569 mandatory capabilities listed above, but may be extended as needs 570 dictate. The CDNI FCI specification SHOULD define the registry (and 571 the rules for adding new entries to the registry) for the different 572 capability types. Each capability type MAY further have a list of 573 valid values. The individual CDNI interface specifications which 574 define a given capability SHOULD define any necessary registries (and 575 the rules for adding new entries to the registry) for the values 576 advertised for a given capability type. 578 The mandatory capabilities listed above generally relate to 579 information that is configured on a content asset or group of assets 580 basis via CDNI metadata. The capability requirements for acquisition 581 and delivery protocol and other mandatory metadata capabilities (e.g. 582 authorization algorithms) are defined in [I-D.ietf-cdni-metadata]. 584 Note: CDNI interface support for logging configuration (i.e., control 585 interface vs. metadata interface) has not yet been decided. Once it 586 has been decided, the corresponding CDNI interface specification 587 should define the associated capability requirements. 589 6. Negotiation of Support for Optional Types of Footprint/Capabilities 591 The notion of optional types of footprint and capabilities implies 592 that certain implementations may not support all kinds of footprint 593 and capabilities. Therefore, any FCI solution protocol must define 594 how the support for optional types of footprint/capabilities will be 595 negotiated between a uCDN and a dCDN that use the particular FCI 596 protocol. In particular, any FCI solution protocol needs to specify 597 how to handle failure cases or non-supported types of footprint/ 598 capabilities. 600 In general, a uCDN may ignore capabilities or types of footprint it 601 does not understand; in this case it only selects a suitable 602 downstream CDN based on the types of capabilities and footprint it 603 understands. Similarly, if a dCDN does not use an optional 604 capability or footprint which is, however, supported by a uCDN, this 605 causes no problem for the FCI functionality because the uCDN decides 606 on the remaining capabilities/footprint information that is being 607 conveyed by the dCDN. 609 7. IANA Considerations 611 IANA registries are to be used for mandatory and optional types of 612 footprint and capabilities. Therefore, the mandatory types of 613 capabilities listed in this document (see Section 5) are to be 614 registered with IANA. In order to prevent namespace collisions for 615 capabilities a new IANA registry is requested for the "CDNI 616 Capabilities" namespace. The namespace shall be split into two 617 partitions: standard and optional. 619 The "standard" namespace partition is intended to contain mandatory 620 to implement capabilities and conforms to the "IETF Review" policy as 621 defined in [RFC5226]. The registry contains the name of the standard 622 capability, the RFC number of the specification defining the 623 capability, and the version number of the FCI capability set to which 624 the standard capability applies. 626 The following table defines the initial capabilities for the standard 627 partition: 629 +----------------------+---------+---------+ 630 | Capability | RFC | Version | 631 +----------------------+---------+---------+ 632 | Delivery Protocol | RFCthis | 1 | 633 | | | | 634 | Acquisition Protocol | RFCthis | 1 | 635 | | | | 636 | Redirection Mode | RFCthis | 1 | 637 +----------------------+---------+---------+ 639 [Ed. Note: Need to add the CDNI Metadata Interface 640 [I-D.ietf-cdni-metadata] and the CDNI Logging Interface 641 [I-D.ietf-cdni-logging] capabilities to this table before 642 publication, or remove this note if this ends up being handled in 643 those documents directly.] 645 The initial FCI version number is set to 1. All three initial 646 capabilities are considered mandatory to implement for version 1. 647 The version field should be incremented when new capability sets are 648 added to the registry. 650 The "optional" namespace partition conforms to the "Expert Review" 651 policy as defined in [RFC5226]. The expert review is intended to 652 prevent namespace hoarding and to prevent the definition of redundant 653 capabilities. Vendors defining new capabilities which conflict with 654 existing capabilities follow the guidelines for the "Specification 655 Required" policy as defined in [RFC5226]. The Version field in the 656 registry is set to "-1" (negative one) for non-standard capabilities. 658 7.1. Footprint Sub-Registry 660 The "CDNI Metadata Footprint Types" namespace defined in the CDNI 661 Metadata Interface document [I-D.ietf-cdni-metadata] contains the 662 supported footprint formats for use in footprint advertisement. No 663 further IANA action is required here. 665 7.2. Protocol Sub-Registry 667 The "CDNI Metadata Protocols" namespace defined in the CDNI Metadata 668 Interface document [I-D.ietf-cdni-metadata] contains the supported 669 protocol values for the Delivery Protocol and Acquisition Protocol 670 capabilities. No further IANA action is required here. 672 7.3. Redirection Mode Sub-Registry 674 The "CDNI Capabilities Redirection Modes" namespace defines the valid 675 redirection modes that may be advertised as supported by a CDN. 676 Additions to the Redirection Mode namespace conform to the "IETF 677 Review" policy as defined in [RFC5226]. 679 The following table defines the initial Redirection Modes: 681 +------------------+----------------------------------+---------+ 682 | Redirection Mode | Description | RFC | 683 +------------------+----------------------------------+---------+ 684 | DNS-I | Iterative DNS-based Redirection | RFCthis | 685 | | | | 686 | DNS-R | Recursive DNS-based Redirection | RFCthis | 687 | | | | 688 | HTTP-I | Iterative HTTP-based Redirection | RFCthis | 689 | | | | 690 | HTTP-R | Recursive HTTP-based Redirection | RFCthis | 691 +------------------+----------------------------------+---------+ 693 8. Security Considerations 695 Security considerations will be discussed in a future version of this 696 document. 698 9. References 700 9.1. Normative References 702 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 703 Requirement Levels", BCP 14, RFC 2119, March 1997. 705 [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an 706 IANA Considerations Section in RFCs", BCP 26, RFC 5226, 707 May 2008. 709 [RFC6707] Niven-Jenkins, B., Le Faucheur, F., and N. Bitar, "Content 710 Distribution Network Interconnection (CDNI) Problem 711 Statement", RFC 6707, September 2012. 713 [RFC6770] Bertrand, G., Stephan, E., Burbridge, T., Eardley, P., Ma, 714 K., and G. Watson, "Use Cases for Content Delivery Network 715 Interconnection", RFC 6770, November 2012. 717 9.2. Informative References 719 [I-D.ietf-cdni-framework] 720 Peterson, L., Davie, B., and R. Brandenburg, "Framework 721 for CDN Interconnection", draft-ietf-cdni-framework-09 722 (work in progress), January 2014. 724 [I-D.ietf-cdni-logging] 725 Faucheur, F., Bertrand, G., Oprescu, I., and R. 726 Peterkofsky, "CDNI Logging Interface", draft-ietf-cdni- 727 logging-09 (work in progress), February 2014. 729 [I-D.ietf-cdni-metadata] 730 Niven-Jenkins, B., Murray, R., Watson, G., Caulfield, M., 731 Leung, K., and K. Ma, "CDN Interconnect Metadata", draft- 732 ietf-cdni-metadata-04 (work in progress), December 2013. 734 [I-D.ietf-cdni-requirements] 735 Leung, K. and Y. Lee, "Content Distribution Network 736 Interconnection (CDNI) Requirements", draft-ietf-cdni- 737 requirements-17 (work in progress), January 2014. 739 Appendix A. Acknowledgment 741 Jan Seedorf is partially supported by the CHANGE project (CHANGE: 742 Enabling Innovation in the Internet Architecture through Flexible 743 Flow-Processing Extensions, http://www.change-project.eu/), a 744 research project supported by the European Commission under its 7th 745 Framework Program (contract no. 257422). The views and conclusions 746 contained herein are those of the authors and should not be 747 interpreted as necessarily representing the official policies or 748 endorsements, either expressed or implied, of the CHANGE project or 749 the European Commission. 751 Jan Seedorf has been partially supported by the COAST project 752 (COntent Aware Searching, retrieval and sTreaming, http://www.coast- 753 fp7.eu), a research project supported by the European Commission 754 under its 7th Framework Program (contract no. 248036). The views and 755 conclusions contained herein are those of the authors and should not 756 be interpreted as necessarily representing the official policies or 757 endorsements, either expressed or implied, of the COAST project or 758 the European Commission. 760 Martin Stiemerling provided initial input to this document and 761 valuable comments to the ongoing discussions among the authors of 762 this document. Thanks to Francois Le Faucheur and Scott Wainner for 763 providing valuable comments and suggestions to the text. 765 Authors' Addresses 767 Jan Seedorf 768 NEC 769 Kurfuerstenanlage 36 770 Heidelberg 69115 771 Germany 773 Phone: +49 6221 4342 221 774 Fax: +49 6221 4342 155 775 Email: seedorf@neclab.eu 777 Jon Peterson 778 NeuStar 779 1800 Sutter St Suite 570 780 Concord CA 94520 781 USA 783 Email: jon.peterson@neustar.biz 785 Stefano Previdi 786 Cisco Systems 787 Via Del Serafico 200 788 Rome 0144 789 Italy 791 Email: sprevidi@cisco.com 793 Ray van Brandenburg 794 TNO 795 Brassersplein 2 796 Delft 2612CT 797 The Netherlands 799 Phone: +31-88-866-7000 800 Email: ray.vanbrandenburg@tno.nl 801 Kevin J. Ma 802 Azuki Systems, Inc. 803 43 Nagog Park 804 Acton MA 01720 805 USA 807 Phone: +1 978-844-5100 808 Email: kevin.ma@azukisystems.com